Electronic and chemical structure of functional oxides

We study oxide band structure, surface and interface chemistry, ferroelectric and multiferroic films.

The defining property of a Ferroelectric (FE) material is a spontaneous macroscopic polarization which can be reversed under an applied electric field. The polarization as a function of applied electric field exhibits a hysteresis loop, analogous to ferromagnetic materials, hence the name ferroelectricity.

It was discovered by Valasek from the University of Minnesota who presented his results at the Washington meeting of the American Physical Society in April 1920.

Perovskite oxides, of general formula ABO3 with a pseudocubic structure, where A and B are two different cations, furnish many interesting ferroelectrics. The B-type cation is octahedrally coordinated with oxygen. In the example shown, BaTiO3, it is the relative symmetry breaking displacement of the Ti atoms with respect to the O atoms which is responsible for the spontaneous polarization. BaTiO3 has three ferroelectric phases: tetragonal, orthorhombic and rhombohedral.

Epitaxial thin films

The advent of high quality epitaxial film growth, decisive theoretical advances and experimental tools capable of characterizing in detail ferroelectric materials has led to a resurgence of interest, in particular the perspective of engineering films for FE-based electronics. Substrate-imposed strain, for example, can increase the Curie temperature, stabilizing ferroelectricity in BaTiO3 up to 600°C.

Ferroelastic domain walls in CaTiO3

Ferroelastic materials display domain structures of uniform strain state separated by very narrow twin walls, typically a few nanometers wide where the spontaneous strain changes sign. Large strain gradients exist in the vicinity of the walls, generating novel properties. At the twin walls in CaTiO3, one of the octahedral tilts goes to zero allowing the emergence of polarity by off-centering of the Ti cation. This could be augmented and controlled electromechanically, adding functionality to the system.

On the way to PHz electronic devices

The discovery of a groundbreaking way to produce 2D nanomaterials and the development of the controllable synthesis of 2D nanomaterials kicked off the experiments exploring the exciting properties of such systems [1]. Nowadays, the family of 2D nanomaterials includes superconductors, zero band-gap semiconductors or semimetals, low band-gap semiconductors, and isolators. In this project, we study the topological properties of monolayer transition metal dichalcogenides (MX2, where M = Mo, W, and X = S, Se, Te). Recently it has been theoretically shown that topological phase transition can be stimulated entirely with shaped light fields, such as trefoil polarization states[2], in conventional hexagonal materials[3]. By orienting the trefoil symmetry it should be possible to populate K and K’ valleys in the conduction band and study the dynamic of electrons using pump-probe photoemission experiments. This will open a route towards new devices based on the electronic topology of 2D materials and nano-structured ultra-short light pulses working in the PHz regime.

Using the microscopic capabilities of nanoESCA, we can select the optimal single flake (size, orientation). k-PEEM allows studying the occupied states in the valence band, while k-SEEM (k-secondary electron emission microscopy) allows studying the unoccupied states. In the future, with an aim to conduct the pump-probe photoemission experiments, the nanoESCA will be upgraded with a special version of sample manipulator allowing illumination of the sample from the backside (pump beam with trefoil polarization).

 1. C. Jin et al., Nat. Nanot. 13, 994–1003 (2018), M. Chhowalla et al., Nature Chemistry 5, 263–275 (2013), B.A. Bernevig et al., Science 314, 1757 (2006)
2. A. Fleischer et al., Nat. Phot.8 (7), 543–49 (2014), L. Rego et al., Science 364 (6447), 1253 (2019)
3. Á. Jiménez-Galán, et al., Nat. Phot. Lett. 14, 728-732 (2021)

Electrical boundary conditions

The new physics emerging from two-dimensional films in the limit of a few unit cells has a host of exciting applications. However, understanding the ferroelectric properties of such engineered thin film systems requires taking into account not only the material but also its interfaces with electrodes, substrates or atmosphere; in other words, the electrical boundary conditions. In the case of a thin film these can even determine the ferroelectric polarization stability. The depolarizing field can place a lower limit on the film thickness capable of supporting a stable polarization.

Multi-ferroïcs

A multi-ferroïc material possesses simultaneously two or more ferroïc orders: ferroelectric, ferromagnetic and ferroelastic. Particularly fascinating is the coupling between these orders. For example magneto-electric coupling allows electrical control of the magnetization or, inversely, magnetic control of the polarization. The coupling coefficient a is defined by M=aP. Multi-ferroicity provides additional handles to control oxide properties. To these can be added collective orbital, spin and charge phenomena.

Interface chemistry

The question of the interface is a key issue in multi-ferroïc heterostructures. Hybridization between filled d orbitals responsible for magnetization and empty d orbitals in the ferroelectric oxide may be one path to such coupling. Several coupling mechanisms have been identified and these can be quite complex. For example, the charge ordering of a magnetic layer can be modulated by the polarization state of an adjacent ferroelectric.

Screening

Surface polarization charge in ferroelectric (FE) materials can be screened by a variety of mechanisms: intrinsic (charge carriers or defects in the bulk), extrinsic (chemical environment or adsorbates), domain ordering, or even a combination of the above. Chemisorption of OH- and protons can lead to important changes in the electrical boundary conditions and water film can play an active role in domain switching. Photo-generated charge carriers can also efficiently screen the surface and interface polarization charge and hence the depolarizing field. The photo-generated pair lifetimes depend on the FE surface topology and the polarization state can influence, for example, redox reactions

Switching the polarization

Switching requires a metallic contact, raising fundamental issues about the interface between the FE and the electrode. Polarization leads to fixed charge of opposite sign at the two metal-FE interfaces. Electrode free charge acts to screen the polarization charge creating dipoles of the same sign at the two interfaces, however, the screening is always imperfect and the residual depolarizing field alters the electrostatic potential inside the FE. The partially covalent nature of the bonds in the FE changes the band structure with respect to that of a perfectly ionic compound which, from consideration of dynamical charge tensors, depends on the atomic distortion. Reduction/oxidation conditions at the surface can control the film polarization.

Response to applied bias

Operational interfaces require knowledge of the interface response to applied bias. The electronic structure near the electrode will be determined by a combination of the above phenomena. For example, bias can induce the formation of filaments of oxygen vacancies or the migration of the latter towards electrodes as in resistive switching, whereas in tunnel regime capacitors atomic distortions can change under bias via the inverse piezoelectric effect.

Phenomenological and first-principles-based theoretical approaches have been deployed to understand the electronic structure at the interface under polarization. Classical semiconductor theory of the metal/insulator interface applied to FE capacitors, including the effect of polarization on band line-up and transport, describes charge injection in terms of the interface whereas the leakage current may be bulk limited. Theory has also shown that the local chemistry at the electrode/FE interface can enhance ferroelectricity. Dissimilar electrodes can induce a built-in electric field in the FE film. Extensive work on FE tunnel junctions has demonstrated the crucial role of the interface chemistry on tunneling probabilities. First principle calculations even predict an ohmic barrier but this may be due to the underestimation of the band gap. Knowledge of the microscopic layer polarization is vital because the interface induced changes of the electronic structure are expected to take place over a 1-2 nanometers.

Domain Walls

Domain walls are intrinsic to ferroic materials. Long considered as understood, it is only recently, thanks to atomic-resolution studies using transmission electron microscopy (TEM) or atomic force microscopies (AFM, c-AFM, PFM etc.) that their real structural and electronic complexity has been revealed. Based on this, a new paradigm of ferroic devices is now envisaged where the domain walls, rather than the domains, are the active element. This field has been coined “Domain boundary engineering” or “Domain wall nanoelectronics”.

Photoferroelectrics are materials with both photoelectric and ferroelectric properties. As correlated-electron materials, they are attracting increasing interest because of their high technological potential. Recent reports show that low conversion efficiencies can be counterbalanced by large, above-band gap photo-voltages in complex multiferroic oxides and the fundamental role of the potential difference across domain walls.

More detailed and quantitative knowledge of its structural, electronic and chemical nature is necessary. Fundamental questions on the crystal structure and the electronic structure in the vicinity of the domain wall, as well as their behavior under illumination remain to be elucidated.

Chemisorption

In view of the potential of transition metal (TM) oxides for photocatalytic applications, the adsorption and dissociation of water on their surfaces has been widely studied. Surface chemistry can strongly influence the adsorption mechanism. For example, the SrO-terminated perovskite oxide SrTiO3 (STO) favors dissociative adsorption, whereas on the TiO2-terminated surface, molecular adsorption is more stable. Chemisorption can be enhanced by precursor-mediated adsorption, where the molecule is first trapped in a weakly bound physisorbed state and then migrates to encounter a more active site, for example, an oxygen vacancy (VO), at which chemical bonds are formed.

The interaction with ferroelectric oxide surfaces is less well-known, yet it is of considerable importance because hydrogen yield can be enhanced by an order of magnitude at the surface of a ferroelectric. The static charge on the surface changes the depth of the physisorption well, which determines the average residence time of the precursor on the surface, leading to a greater chance of finding a chemisorption site.

Another important aspect of dissociative adsorption is the production of H+ capable of bonding to lattice oxygen. The presence of hydrogen in ferroelectrics is known to be a factor that increases imprint, potentially inhibiting switching of the polarization. In PbTiO3, OH− adsorption favors an upward pointing polarization state, whereas H+ adsorption stabilizes downward pointing polarization.